Notes for Data Users

1997 Data Limitations

For data year 1997, 29 states used the Mitofsky-Waksberg method of sampling, 17 states used a disproportionate stratified sample (DSS) design, and 6 states used a variety of other designs. The BRFSS standard for sample designs is that sample records must be justifiable as a probability sample of all households with telephones in the state. All but Alaska, California, Nevada and Texas met this criterion in 1997. The deviations from the standard by these states may reduce the comparability of their data with those of other areas.

Alaska used a stratified directory/random sample combination where the sample design is made up of 3 random digit-dialed strata and 4 listed strata. Telephone numbers in the listed strata are selected from a list of putative household telephone numbers. Forty-five percent of the completed interviews come from the listed strata. Alaska's sampling frame for the listed strata contains an estimated 50% of all household telephone numbers in these strata.

California used a truncated list-assisted design with sampling from 3+ blocks only. Noncoverage was further affected to an unknown degree by substitution for sample numbers used by any client of the commercial sample provider in the preceding six months. Probabilities of selection varied in proportion to the number of listed household telephone numbers in the 100 blocks associated with a number, the number of known business telephone numbers preceding a number, the number of preceding telephone numbers selected by any client of the commercial sample provider in the preceding six months, and the measure of size used for each county, which could vary each time a sample was drawn.

Nevada used a truncated list-assisted design with sampling from 5+ blocks only. Noncoverage was further affected to an unknown degree by substitution for sample numbers used by any client of the commercial sample provider in the preceding six months. Probabilities of selection varied in proportion to the number of listed household telephone numbers in the 100 blocks associated with a number, the number of known business telephone numbers preceding a number, the number of preceding telephone numbers selected by any client of the commercial sample provider in the preceding six months, and the measure of size used for each county, which could vary each time a sample was drawn

California modified the wording of mammography, Papanicolaou (PAP) test, and chronic alcohol use and is therefore excluded from those tables in the report. California also asked the HIV/AIDS section questions to persons 18-44 years of age rather than those 18-64 years of age, and is therefore excluded from tables reporting data on this topic.

Although states collect data monthly, the number of interviews ranged from 1500 in New Hampshire to 4896 in Idaho in 1997. Estimates for some states are therefore based upon smaller sample sizes, and for rare events (e.g., diabetes), may yield unstable estimates.

Notes for Data Users library  2002 | 2001 | 2000 | 1999 | 1998 | 1997 | 1996

Policies and Regulations | BRFSS Home | Contact Us

CDC Home | Search | Health Topics A-Z

This page last reviewed June 22, 2005

United States Department of Health and Human Services
Centers for Disease Control and Prevention
National Center for Chronic Disease Prevention and Health Promotion
Division of Adult and Community Health